Computer Engineering Commons^™

A Comprehensive Framework For Targeted Transfer Learning: Privacy-Preserving, Secure, And Efficient Domain Adaptation, Sayyed Farid Ahamed

Electrical & Computer Engineering Theses & Dissertations

Machine learning models deployed in tactical environments rely on high-quality labeled datasets, which are often challenging, expensive, and time-consuming to obtain. In addition, formal techniques for assuring model performance under dynamic conditions remain limited. Transfer learning has emerged as a promising solution, but its effectiveness is limited by domain coverage and similarity between source and target datasets. This dissertation presents a comprehensive, modelagnostic Targeted Transfer Learning (TTL) framework that addresses transferability challenges between domains with minimal data, demonstrating how targeted sample selection improves adaptation across image classification, network security, and privacy-preserving collaborative settings. First, we develop the TTL framework for …

Physics-Guided Deep Learning For Predictive Modeling Of Spatiotemporal Dynamical Systems, Niharika Deshpande

Engineering Management & Systems Engineering Theses & Dissertations

Many physical and networked systems evolve under continuously changing spatial and temporal conditions. Transportation networks respond to fluctuating demand, atmospheric fields reorganize as storms intensify, and coastal response depends on localized forcing pathways. Modeling such systems requires learning formulations that adapt to evolving structure, operate on irregular geometries, and provide interpretable measures of predictive uncertainty. This dissertation develops a physics-guided spatiotemporal learning framework designed for structured dynamical systems whose governing interactions are neither static nor Euclidean. The central premise is that spatial relationships in these systems are dynamic and geometry-dependent. To represent this behavior, system states are modeled on time-varying …

Towards More Realistic And Practical Graph Backdoor Attacks, Jiawei Chen

Computer Science Theses & Dissertations

Graph Neural Networks (GNNs) have demonstrated remarkable performance on graph-based learning tasks and are increasingly deployed in security-critical applications. However, recent studies have shown that they are highly vulnerable to graph backdoor attacks (GBAs), where adversaries implant malicious triggers to induce targeted misclassification during inference. Despite their effectiveness, existing GBAs are often developed under unrealistic assumptions, such as focusing exclusively on simple homogeneous graphs or assuming the adversary possesses privileged access to target nodes during inference. This dissertation aims to systematically investigate and design graph backdoor attacks under significantly more realistic graph settings and adversarial constraints.

First, we investigate the …

Enhancing Stem Education With Modeling, Simulation, And Ai Technologies: From Virtual Laboratories To Intelligent Teaching Assistants, Yiyang Li

Electrical & Computer Engineering Theses & Dissertations

Rapid advancements in modeling and simulation (M&S) and artificial intelligence (AI) present new opportunities to enhance various aspects of STEM education, from virtual laboratories that simulate physical lab environments in software to intelligent teaching assistants that provide on-demand, curriculum-aligned instructional support. Virtual laboratories offer a potential solution to the access and scalability challenges of laboratory courses by allowing students to conduct experiments without physical equipment or geographical constraints. AI-powered teaching assistants, particularly those grounded in course-specific materials, can help mitigate the instructional support gap that arises when students work independently in digital learning environments. This dissertation presents three-phase research into …

Deep Multimodal Sensor Fusion For Autonomous Driving: A Real-World Deployment On Little Blue, Raphael T. Pamie-George

Electrical & Computer Engineering Theses & Dissertations

Accurate perception and control systems are essential for autonomous vehicles to safely interpret and navigate complex environments. Individual sensing modalities such as cameras, LiDAR, and inertial measurement units (IMUs) provide valuable environmental and vehicle state information; however, each sensor exhibits inherent limitations when used independently. Camera performance can degrade under varying lighting conditions, LiDAR measurements may be sparse or affected by environmental noise, and IMU measurements are susceptible to drift over time. Little Blue, a research autonomous vehicle platform from ODU (Old Dominion University), previously relied primarily on camera-based data for neural network training, limiting the system’s ability to exploit …

Parameter Dependent Chen-Fliess Series And Their Nonrecursive Interconnections, Natalie T. Pham

Electrical & Computer Engineering Theses & Dissertations

In control theory, a Chen-Fliess functional series is a weighted sum of iterated integrals constructed from a given set of input functions. Such series can be used to represent nonlinear input-output systems. In applications, they have been employed to characterize interconnected nonlinear systems, to solve system inversion and tracking problems, and to design predictive and adaptive controllers.

Distributed parameter systems exhibit spatial dependence along with temporal dependence. Such systems are typically represented in terms of partial differential equations. In control theory, there appears to be no existing method for representing the input-output map of a distributed system via a Chen-Fliess …

Advancing Imbalanced Classification Through Representation Learning, Morteza Mohammady Gharasuie

Computer Science Theses & Dissertations

Real-world datasets frequently exhibit severe class imbalance, where certain categories are significantly underrepresented relative to others, leading standard empirical risk minimization to bias learning toward majority classes and degrade performance on minority categories. This dissertation addresses imbalanced classification across both structured tabular data and long-tailed visual recognition by developing methods that improve representation learning and decision reliability under skewed distributions. For tabular data, it introduces Conditional Probability Representation (CPR), a target-based encoding framework that embeds feature–label relationships directly into the representation space, enhanced by a progressive feature upgrading mechanism for semi-supervised settings and a class-frequency–aware extension that improves robustness to …

Improving Fairness On Semantic Segmentation Using Large Language Models, Samuel E. Burggraf

Electrical & Computer Engineering Projects for D. Eng. Degree

As machine learning systems are increasingly integrated into critical decision-making processes, ensuring fairness in their design and implementation has become a significant concern. While fairness research has primarily focused on specific protected attributes, less attention has been given to spatial fairness, which can affect individuals at specific locations. If fairness is not addressed, models may systematically underperform in certain regions or across populations which can lead to unequal access to accurate predictions and potentially biased decision-making. Fairness considerations should extend across all machine learning applications to align with the National Institute of Standards and Technology (NIST) guidelines of fair and …

Landscaping Of Mcp: An Overview Of Mcp Mitigations And Tools, Arden Michel

Cybersecurity Undergraduate Research Showcase

The Model Context Protocol (MCP) has quickly become the standard for enabling agentic AI systems to interact with external tools, data sources, and services. Since its debut in 2024, MCP has been adopted by companies such as Google, Apple, Meta, and IBM. While this integration greatly improves the capabilities of large language models (LLMs), it also creates a new attack surface that the security community is only beginning to understand systematically.

A key architectural challenge is MCP's fundamental reliance on implicit trust: servers often run locally with high privileges, tool descriptions are accepted without question, and external servers are presumed …

Limitations Of Signature-Based Network Intrusion Detection Under Modern Traffic Conditions, Henry Guidry

Cybersecurity Undergraduate Research Showcase

Network Intrusion Detection Systems are tools used to monitor network traffic and alert to suspicious or harmful activity before it can cause harm. Signature-based versions of these systems are a foundation for intrusion detection, operating by finding common patterns and forming malicious signatures. However, three developments in modern network environments have greatly impacted the significance of Network Intrusion Detection Systems. These three developments are the near-complete adoption of end-to-end encryption, the use of sophisticated packet fragmentation techniques, and the processing demands of high-throughput networks. Encryption makes deep packet inspection practically infeasible by transforming inspectable payloads into ciphertext, forcing NIDS to …

Escaping Isolation: An Analysis Of Virtual Machine And Container Breakout Vulnerabilities, Felix Iov

Cybersecurity Undergraduate Research Showcase

Cloud computing providers rely on multi-tenant architectures to maximize resource efficiency. This infrastructure depends on virtualization, which provides isolation between clients. This comes primarily in the form of Virtual Machines (VMs) and Containers. However, “breakout attacks” or “escapes” are a critical threat where attackers bypass these isolation layers to gain unauthorized access to the host system and neighboring environments. This paper surveys virtualization escape threats and analyzes three case studies: a runc container escape (Leaky Vessels), a VMware ESXi VM escape (VSOCKPuppet), and an NVIDIA GPU container escape (NVIDIAScape). Each demonstrates different attack surfaces, including file descriptor misuse, kernel driver …

The Psychology Behind Ai-Generated Phishing And Social Engineering Attacks, A’Shya Reynolds

School of Cybersecurity Master's Level Projects and Papers

Cybercrime has evolved significantly with the integration of artificial intelligence (AI), transforming traditional phishing and social engineering attacks into highly sophisticated and personalized threats. While early phishing attempts relied on generic messaging and low success rates, modern AI-driven attacks leverage advanced data analytics, natural language processing, and behavioral prediction to manipulate victims more effectively.

This research examines how cybercriminals utilize AI to enhance psychological manipulation techniques in phishing and social engineering attacks, increasing victim susceptibility. Drawing from interdisciplinary literature in cybersecurity and psychology, this study explores key psychological mechanisms, including cognitive biases, emotional triggers, and decision-making processes that influence victim …

An Integrated Bayesian Network-Based Zero Trust Model To Quantify Cyber Risk In Small-Medium Businesses, Ahmed Abdelmagid

Engineering Management & Systems Engineering Theses & Dissertations

Small-medium businesses (SMBs) play a pivotal role in the worldwide economy as they constitute the most considerable portion of businesses in developed countries like the UK and the US. As such, SMBs are likely targets of cybercrimes by malicious agents because of their vulnerable IT systems. The digital infrastructure of SMBs is more likely to be hit by cyberattacks than large businesses due to many factors that facilitate hackers’ missions. These factors include a limited financial budget devoted to cybersecurity, a lack of knowledge, an underrating of how dangerous cyber threats are, and a shortage of IT expertise. The enormous …

Influence Of Gender-Specific Data Imbalance On Scgpt Fine-Tuning For Single-Cell Genomics, Mohammad Aman Ullah Al Amin, Daniil Filienko, Hong Qin

The Knowledge and Creativity Expo

The transformer-based foundation model scGPT has demonstrated strong capabilities in analyzing high-dimensional single-cell RNA sequencing data. However, the impact of demographic factors, particularly gender, on model performance remains insufficiently understood. Gender is known to influence cell-type compositions in the immune system. Here, using the gender-sensitive cell-type composition in immune system, we comprehensively evaluated how the gender-sensitive imbalance of training data influences the performance of scGPT in cell-type predictions. We fine-tuned scGPT on male-only, female-only, and mixed-gender subsets from two large-scale datasets containing immune cells. We used a logit difference to measure the confidence gap between the true label and the …

Viability Of Widely Used Encryption Schemes In Drone Transmission, Emanuel Yasir Nelson

Cybersecurity Undergraduate Research Showcase

This paper presents throughout research on the security issues related to drone transmission. These topics were addressed and explained, in particular the aspects relating to cybersecurity, for utmost clarity. These include threats and vulnerabilities, drone transmission the impact of encryption on latency, and the details of the encryption methods AES-128, AES-256, and ChaCha20 that were used in the experiment described in the paper. Each encryption method performance was measured and outputted by the Python code developed and used in the experiment. Afterwards, the performance of each method was analyzed in relation to their decryption time, encryption time, end to end …

A Transition Framework For Hybrid Tls In Enterprise-Level Systems, William Hadd

Cybersecurity Undergraduate Research Showcase

Enterprises face an immediate need to protect long-lived data against harvest-now, decrypt-later threats while maintaining interoperability across layered systems. With NIST’s first post-quantum standards finalized (ML-KEM, ML-DSA, SLH-DSA) and TLS hybridization drafts defining concrete ECDHE + ML-KEM groups, adoption can begin at the TLS termination layer even before full ecosystem support for post-quantum signatures arrives (NIST, 2024; IETF, 2025). In this paper, we propose an enterprise-oriented transition framework and maturity model for hybrid TLS across email, internal API gateways, and object storage. We specify where to enforce, which hybrid groups to select, and how to prevent silent downgrade with policy …

Bridging The Gap Between Network Science And Network Systems To Identify And Mitigate Cyber Risk: Identify And Mitigate Backdoor Attacks On Graph Neural Networks And On Complex Systems, Sabah Ettahri

Electrical & Computer Engineering Projects for D. Eng. Degree

This doctoral project aims to bridge the gap between graph theory and network science to identify and mitigate cyber risk, represented as a CY-Triangular Network that connects different networks. The CY-Triangular Framework is a cybersecurity system that integrates graph theory and network science through an interoperable learning approach. The objective of this project is to bridge the gap between two domains: network science and network systems. Accordingly, it examines one representative network from each field, focuses on a complex system network, and explores Graph Neural Networks (GNNs). The connection between these domains lies in graph theory. This research demonstrates that …

Human Identification And Action Recognition Using Small Data And Deep Domain Adaptation, Alexander M. Glandon

Electrical & Computer Engineering Theses & Dissertations

Human identification and human action recognition problems are two important research areas for real-world security and surveillance applications. In both human identification and action recognition, it is necessary to operate by collecting small datasets in the field, possibly in a short time window of observation. This dissertation studies and develops computational modeling and high-performance machine learning (ML) and deep learning (DL) models for human identification and human action recognition using small amounts of data. These methods and computational models may be useful for different security and surveillance applications.

This dissertation on human recognition develops a ML computational model to estimate …

Machine Learning For Anomaly Detection In Neural Network Security And Srf Cavities, Hal Ferguson

Electrical & Computer Engineering Theses & Dissertations

This dissertation explores the development and deployment of machine learning approaches to address critical challenges in anomaly detection across two distinct domains: neural network security in federated learning settings and cavity behavior analysis in particle accelerator operations at Jefferson Lab in Newport News, Virginia. Anomaly detection identifies deviations from expected patterns, safeguarding systems in cybersecurity, industry, and research against malicious activities and failures. This dissertation demonstrates how our machine learning approaches enhance detection accuracy and efficiency in both neural network security and industrial applications.

First, we investigate vulnerabilities in deep neural networks deployed in federated learning. Although federated learning preserves …

Multimedia Forensics: Identification And Verification Of Source Camera, Vehicle Speed Estimation, And Deepfakes Detection, Jiajun Jiang

Electrical & Computer Engineering Theses & Dissertations

This dissertation advances multimedia forensics by addressing three critical research areas that enhance the authenticity verification and analysis of digital media. Multimedia forensics, which encompasses techniques for examining images, videos, audio, and text, faces increasing challenges due to sophisticated editing tools and massive data volumes. In the first study, a fast source camera identification and verification method based on PRNU analysis is proposed for video forensic investigations. By integrating camera rolling and I-frame analysis, this approach achieves a processing speed improvement of at least 15 times over conventional frame-by-frame methods while reducing false positives. The second study focuses on vehicular …

Computational Modeling For Automatic Superconducting Cavity Fault Prediction And Classification Using Time Series Signals, Md Monibor Rahman

Electrical & Computer Engineering Theses & Dissertations

Processing multivariate time series signals collected from sensor networks is challenging because of complex temporal dependencies and non-stationarity. With the advent of artificial intelligence (AI) like machine learning and deep learning, it has become possible to process sensor-driven time series data more effectively than traditional statistical methods.

This dissertation aims to develop machine learning and deep learning models to address machine fault diagnosis using multivariate time series signals collected from the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. The first goal of the proposed work is to develop deep learning–based classification models and an unsupervised fault clustering approach …

Human Activity Recognition And Identification Driven Automated Deep Learning For Time-Series Classification, Justin Alan Gamble

Engineering Management & Systems Engineering Theses & Dissertations

The growing emphasis on Digital Engineering (DE) within the U.S. Department of Defense (DoD) demands advanced methods for leveraging vast time-series data generated by sensor-rich environments. Deep learning models offer promising solutions for complex timeseries classification tasks, however their design and optimization remain highly resource intensive, requiring specialized expertise. This dissertation addresses this challenge by developing and evaluating an Automated Machine Learning (AutoML) framework specifically tailored for the time-series classification task of Human Activity Recognition and Identification (HARI).

A systematic investigation was conducted using the Design Science Research Methodology (DSRM) comparing traditional search strategies of grid search and random search …

Input Structure Based Optimization For Privacy Preserving Ai Systems, Feng Yizhou

Electrical & Computer Engineering Theses & Dissertations

As Artificial Intelligence (AI) systems become increasingly integrated into critical domains, ensuring privacy-preserving model design and system deployment has become a pressing priority. Safeguarding both sensitive user data and proprietary model parameters is critical throughout the AI model and system, from data acquisition and pre-processing to model inference and deployment. However, existing privacy-preserving frameworks face several limitations, including fragmented data ownership, incomplete protection across system stages, substantial computational overhead, and poor scalability to modern architectures such as large language models. This dissertation explores a unifying optimization strategy centered on input structure design to address these challenges. The core idea is …

Towards Efficient Privacy-Preserving Deep Learning: He-Friendly Structures, Flexible Pruning, He-Efficient Architectures, And Secure Transformer Token Drop, Yifei Cai

Electrical & Computer Engineering Theses & Dissertations

Deep learning (DL) has become a powerful tool for solving complex problems, but developing DL models typically requires vast datasets, high computational resources, and expert knowledge—barriers that limit accessibility. Machine Learning as a Service (MLaaS) addresses this challenge by allowing resource-rich providers to deliver pre-trained DL models as services. However, privacy concerns arise: clients hesitate to share sensitive data, while providers protect their proprietary models. To address this, privacy-preserving MLaaS integrates cryptographic techniques into DL computations, as seen in frameworks like Cryptonets, SecureML, GAZELLE, CrypTFlow2, Cheetah, and BOLT. Among them, Homomorphic Encryption (HE) enables computation on encrypted data but remains …

Frontlines Of Influence: State Vs. Nonstate Disinformation Campaigns, Lily Wershbale

Cybersecurity Undergraduate Research Showcase

This paper examines the evolution of disinformation campaigns conducted by state and nonstate actors, focusing specifically on Russia and the Islamic State as representative case studies. Through historical examples and qualitative comparative analysis, this research identifies the similarities and differences in disinformation’s role in actors’ core missions, resource allocation, and targeting decisions. This paper further explores the implications of artificial intelligence on disinformation campaigns, investigating how emerging technology will impact the influence operations of state governments and nonstate organizations alike. The findings reveal that while the ultimate intent of both state and non-state actors is to destabilize societies, their approaches …

Digital Thread: Bridging Macro–Micro Services In System-Of-Systems, Joshit Mohanty

Graduate Student Government Association Research Conference

Organizations and industries increasingly rely on distributed services in decentralized environments—ranging from large-scale, system-of-system architectures to fine-grained, agent-based microservices. While this distributed paradigm offers flexibility and innovation, it presents critical challenges such as interoperability gaps, inconsistent data formats, and a lack of holistic oversight. Traditional integration approaches, including ad-hoc middleware or enterprise service buses, tend to solve these issues reactively. As a result, technical debt accumulates, stakeholder misalignments persist, and scaling to new demands becomes complex.

This research proposes digital thread (DT) as the unifying framework to create an authoritative source of truth: a continuous flow of information across the …

Meta-Clustering For Specialized Language Models: Enhancing Contextual Adaptation And Mitigating Hallucinations In Diverse Healthcare Environments, Joshit Mohanty, Vaishali Vaishali, Sandeep Kumar Nayak, Sumit Lahiri

Graduate Student Government Association Research Conference

Large Language Models (LLMs) have significantly advanced conversational AI by enabling dialogic information-seeking and task execution across diverse domains. However, their extensive parameters and broad domain scope lead to “data hallucinations.” These shortcomings are particularly evident in dynamic and diverse environments like India’s healthcare sector, where myriad languages, regional practices, and cultural nuances demand specialized, localized expertise rather than one-size-fits-all generalist models. This paper introduces a meta-clustering framework that integrates Distilled Language Models (DLMs) and Small/Specialized Language Models (SLMs) with meta-learning principles to address these limitations. By drawing on evidence from works such as MedHalu and Med-HALT, the framework seeks …

Why Do Different Llms Give Different Answers To The Same Question? Model Uncertainty And Variability In Llm-Based Intrusion Detection Systems Ranking, Charlise Calloway

Cybersecurity Undergraduate Research Showcase

Large Language Models (LLMs) are increasingly applied across business, education, and cybersecurity domains. However, LLMs can yield varied outputs for the same query due to differences in architecture, training data, and response generation mechanisms. This paper examines model variability and uncertainty by comparing the responses of three LLMs—ChatGPT-4o, Gemini 2.0 Flash, and DeepSeek-V3--to a query on ranking practical intrusion detection systems (IDS). The analysis highlights key similarities and differences in the models’ outputs, offering insight into their respective reasoning and consistency.

Cybermapping Solutions: A Unified Approach In Us/Nato Military Applications And Development, Nicholas Macrino

Electrical & Computer Engineering Projects for D. Eng. Degree

[First paragraph] Cyber threats are evolving in complexity and frequency, posing significant challenges for cybersecurity professionals in identifying, categorizing, and responding to attacks in real time. Unlike traditional warfare, where battlefield awareness is based on fixed geographic warfare, cyber operations involve abstract attack vectors, non-linear threat escalation, and rapidly changing network conditions. Modern cyber threats, such as advanced persistent threats (APTs), polymorphic malware, and distributed denial-of-service (DDoS) attacks, require adaptive visualization techniques that provide real-time awareness and facilitate rapid decision-making. However, existing symbology standards, such as MIL-STD-2525D, were not designed to accommodate the dynamic nature of cyber warfare. The inability …

Testing Autonomy: Hybrid Scenario Synthesis, Benjamin E. Hargis

Electrical & Computer Engineering Theses & Dissertations

Hybrid Scenario Synthesis merges static and adaptive techniques to generate interactions that rigorously assess autonomous performance under multi-factor testing. Multifactor scenarios employ multiple individual stimuli to rigorously test system responses in complex settings. Static Scenario Testing involves scripted test cases that simulate specific conditions or events. These scenarios represent typical situations an autonomous system might encounter. The benefits of static testing include early defect detection, focused review by trained experts, and efficiency. In multi-factor scenarios, however, statically defined scenario factors are not able to guarantee meaningful interactions as the presence of other factors may invalidate underlying assumptions regarding the system …